- Home
- Compare GPUs
- NVIDIA H20 vs NVIDIA H100
NVIDIA H20 vs NVIDIA H100
Side-by-side comparison of specifications, performance, and efficiency.
NVIDIA H20
Architecture
Hopper
Year
2025
VRAM
141 GB
TDP
400 W
NVIDIA H100
Architecture
Hopper
Year
2023
VRAM
94 GB
TDP
400 W
Specifications
NVIDIA H20 NVIDIA H100
Architecture
Hopper Hopper
Launch Year
2025 2023
Form Factor
SXM PCIe
VRAM
141 GB94 GB
Memory Bandwidth
4.0 TB/s3.9 TB/s
TDP
400 W400 W
Process Node
4nm 4nm
| Spec | NVIDIA H20 | NVIDIA H100 |
|---|---|---|
| Architecture | Hopper | Hopper |
| Launch Year | 2025 | 2023 |
| Form Factor | SXM | PCIe |
| VRAM | 141 GB | 94 GB |
| Memory Bandwidth | 4.0 TB/s | 3.9 TB/s |
| TDP | 400 W | 400 W |
| Process Node | 4nm | 4nm |
Performance (TFLOPS)
NVIDIA H20 NVIDIA H100
FP64
1.0 TFlops30.0 TFlops
FP32
44.0 TFlops60.0 TFlops
FP16
148.0 TFlops835.5 TFlops
BF16
148.0 TFlops835.5 TFlops
FP8
296.0 TFlops1670.5 TFlops
INT8
296.0 TFlops1670.5 TFlops
| Precision | NVIDIA H20 | NVIDIA H100 |
|---|---|---|
| FP64 | 1.0 TFlops | 30.0 TFlops |
| FP32 | 44.0 TFlops | 60.0 TFlops |
| FP16 | 148.0 TFlops | 835.5 TFlops |
| BF16 | 148.0 TFlops | 835.5 TFlops |
| FP8 | 296.0 TFlops | 1670.5 TFlops |
| INT8 | 296.0 TFlops | 1670.5 TFlops |
FLOPS by Precision
Efficiency — TDP vs FP32 Performance
Analysis
Most Efficient
NVIDIA
H100
0.15 TFLOPS/W
Where to Rent
NVIDIA H100
| Provider | Instance | Price/hr | |
|---|---|---|---|
RunPod | H100 NVL | $2.59 | View → |
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.